JD.com Scraper: Prices, Review Counts & Ratings, No Login

Type keywords or paste JD.com (京东) product links, and get JD products as one row each: price in CNY, shop, 自营 (JD self-operated) flag, number of reviews, good rate in percent, images, categories and links. No JD login, no cookies, no API key. The prefilled search (手机, 10 products) finishes in about 4 seconds and costs $0.035 on the Free plan.

Use it for price research on the Chinese market, competitor and brand tracking on JD, and product research for cross-border trade. Every search row has the review count and good rate filled when JD shows them.

Table of contents

• What it does
• Quick start
• Use cases
• Input
• Output
• Pricing and limits
• API, export and integrations
• Troubleshooting
• FAQ
• Related Actors
• Support

What it does

The Actor reads public JD.com pages in one of two modes:

• Search — for each keyword, it reads JD's public search results (京东热卖, re.jd.com) page by page. One row per product: price, shop, self-operated flag, review count (5万+) and the same as a number (50000), good rate (99 = 99% positive), main image, category IDs and the position in the results.
• Product — for each JD product link or SKU number, it reads the public mobile product page: title, brand, shop, all images, categories and stock. Then it looks for the same SKU in JD search to add the price, review count and good rate. In our tests on real products the price was found for about 80–90% of them. JD Worldwide (京东国际, jd.hk) products work too.

Every row has status, error, a link and fetchedAt, so you can filter, sort and join rows in the Dataset, a spreadsheet or by API.

Good fit

• Find products by Chinese keyword and compare price, shop and review numbers.
• Track prices of the same keywords over time (run on a schedule and compare by skuId).
• Check how popular a product is: commentsCountMin and goodRate for each SKU.

Not a fit / limits

• About 150–200 products per keyword. JD's public search stops there; later pages repeat. For more products, add more specific keywords (华为手机, 小米手机).
• No review texts, no variant or specification lists, no seller ratings. These JD pages need a login.
• In Product mode the price is best-effort: JD hides prices on product pages without a login, so the Actor finds the price only when the same SKU appears in JD search for its title. When it does not, the row is partial with title, shop and images, and price: null.
• The order of results can differ from the JD app or search.jd.com. Prices are the ones JD shows to a visitor without an account (no member or coupon prices).

Quick start

1. Open the Input tab.
2. Keep Mode = Search by keyword and the prefilled keyword 手机 in Keywords, or type your own (Chinese works best).
3. Keep Max products per keyword = 10 (prefill) or raise it.
4. Click Start.
5. Open the Output tab, view Products, and check price, commentsCount and goodRate.

Expected result: 10 rows with status complete, one per product, in JD search order.

For product links, set Mode = Look up product URLs / SKUs and add links or SKU numbers to Product URLs or SKUs.

Use cases

Buyer job	Input path	Output entity	Key fields	Next integration
Price research on JD	keywords	product	price, shop, selfOperated	Google Sheets, CSV
Popularity and rating check	keywords or productUrls	product	commentsCountMin, goodRate	spreadsheet, BI
Price tracking	keywords + schedule	product	skuId, price, fetchedAt	Dataset API, webhooks
Enrich a list of SKUs	productUrls (mode product)	product	title, brand, shop, images, inStock	your catalog

Input

Input fields

Field	Plain name	Type	Required	Default / prefill	Allowed values	Effect	Example	Recommendation
mode	Mode	string	no	default search	search, product	What to collect	"product"	Use product for a list of known SKUs
keywords	Keywords	array of strings	yes in search	prefill ["手机"]	1–20 keywords, up to 100 characters each	One search per keyword	["笔记本电脑", "耳机"]	Chinese words give the best results; repeated keywords are read once
productUrls	Product URLs or SKUs	array of strings	yes in product	—	up to 50 item.jd.com / item.m.jd.com links or SKU numbers	Products to look up	["100309589315"]	Wrong entries become free error rows
maxItems	Max products per keyword	integer	no	default and prefill 10	1–1000	Most products per keyword	200	JD gives about 150–200 per keyword
proxyConfiguration	Proxy	object	no	{"useApifyProxy": true}	Apify Proxy settings	Proxy for JD requests	{"useApifyProxy": true}	Keep the default

Empty input in the selected mode stops the run with a message that names the field (INVALID_INPUT). Extra spaces are removed and duplicates are read once.

Interacting settings

When	Input A	Input B	Effective behavior	Example	Cost/result effect
Search mode	mode = search	productUrls	Links are ignored	—	None
Product mode	mode = product	keywords, maxItems	Keywords and maxItems are ignored	—	One row per link
Same SKU twice	productUrls	—	Read once	["100309589315", "https://item.jd.com/100309589315.html"]	Charged once

Input examples

Search products by keyword

{
  "mode": "search",
  "keywords": ["手机"],
  "maxItems": 10
}

Expected behavior: reads the first JD search results page for 手机. Expected output: 10 product rows with price, shop, commentsCount, goodRate. Limit or cost note: about 4 seconds, $0.035 on the Free plan.

Look up products by link or SKU

{
  "mode": "product",
  "productUrls": [
    "https://item.jd.com/100309589315.html",
    "100298246508"
  ]
}

Expected behavior: reads each product page, then looks for the SKU in JD search to add the price. Expected output: one row per product with title, brand, shop, images, inStock, and price when found. Limit or cost note: about 10–25 seconds for a few products, about 80 seconds for 50; each product row is charged once, not_found and error rows are free.

Output

The Dataset has one view, Products. The run summary is in the Key-value store record RUN_SUMMARY.

Output fields

Field	Meaning	Type	Can be empty?	Example / enum	Source / notes
skuId	JD SKU number	string	on error rows	"100298246508"
title	Product title	string	on not_found / error	"一加Ace 6T 12GB+256GB 黑色…"	As on JD, in Chinese
price	Price	number	yes	3389	In CNY; null = JD did not show a public price
currency	Currency	string	no	"CNY"	Always CNY
shop	Shop name	string	yes	"一加手机京东自营旗舰店"
shopId	Shop ID	string	yes	"1000001947"
selfOperated	Sold by JD itself (京东自营)	boolean	yes	true
brand	Brand	string	yes	"华为（HUAWEI）"	Product mode only; null in search
commentsCount	Review count as JD shows it	string	yes	"5万+"	万 = 10,000
commentsCountMin	Review count as a number	integer	yes	50000	Lower bound: "5万+" → 50000
goodRate	Positive reviews, percent	integer	yes	99	null = no reviews yet (often new products)
url, mobileUrl	Product page, desktop and mobile	string	no	"https://item.jd.com/100298246508.html"
image, images	Main image, all images	string, array	yes / empty	image URLs	Search: 1 image; product: all images
categoryIds	JD category IDs, level 1–3	array	empty	["9987", "653", "655"]
inStock	In stock	boolean	yes	true	Product mode only
keyword, position, totalResults	Keyword, position in results, results JD reports	string, integer, integer	product mode	"手机", 1, 7680	Search mode only
input	Your input entry	string	search mode	"100298246508"	Product mode only
mode	Mode of the run	string	no	"search"
status	Row status	string	no	complete, partial, not_found, error	See below
error	What is missing and why	string	yes	null	Empty on complete rows
fetchedAt	When the row was read	string	no	"2026-10-15T09:51:01.331Z"	ISO 8601, UTC

Output entities and surfaces

Situation	Dataset/API/KVS/live view	Result shape	User action
Normal item	Dataset, view Products	one row per product, status: complete	Open, export or use the fields
Price not found (product mode)	Dataset	status: partial, price: null, error explains	Use title, shop and images, or search by keyword
SKU does not exist	Dataset	status: not_found, free	Check the number
Wrong link	Dataset	status: error, skuId: null, free	Fix the link
Keyword without products	RUN_SUMMARY.targets	no rows, keyword not_found; RUN_SUMMARY.status is partial (or not_found when no keyword found anything)	Try other words
JD blocks or fails	RUN_SUMMARY.reason: SOURCE_ERROR	rows saved before the failure stay	Run again later

Output example

Search mode, from the prefilled input (one row):

{
  "skuId": "100298246508",
  "title": "一加Ace 6T 12GB+256GB 黑色 oppo 第五代骁龙 8 165Hz 超高刷电竞屏 学生游戏电竞智能手机 国家补贴",
  "price": 3389,
  "currency": "CNY",
  "shop": "一加手机京东自营旗舰店",
  "shopId": "1000001947",
  "selfOperated": true,
  "brand": null,
  "commentsCount": "5万+",
  "commentsCountMin": 50000,
  "goodRate": 99,
  "url": "https://item.jd.com/100298246508.html",
  "mobileUrl": "https://item.m.jd.com/product/100298246508.html",
  "image": "https://img14.360buyimg.com/n0/jfs/t1/426075/25/1266/29522/69ead4abF0cc818dc/0083320320f82202.jpg",
  "images": ["https://img14.360buyimg.com/n0/jfs/t1/426075/25/1266/29522/69ead4abF0cc818dc/0083320320f82202.jpg"],
  "categoryIds": ["9987", "653", "655"],
  "inStock": null,
  "keyword": "手机",
  "position": 1,
  "totalResults": 7680,
  "input": null,
  "mode": "search",
  "status": "complete",
  "error": null,
  "fetchedAt": "2026-10-15T09:51:01.331Z"
}

Product mode (one row, images trimmed):

{
  "skuId": "100309589315",
  "title": "HUAWEI Pura X View 12GB+512GB 亚麻灰 华为阔直屏手机 沉浸阔直板 超轻薄长续航 新品",
  "price": 6999,
  "currency": "CNY",
  "shop": "华为京东自营旗舰店",
  "shopId": "1000004259",
  "selfOperated": true,
  "brand": "华为（HUAWEI）",
  "commentsCount": "2万+",
  "commentsCountMin": 20000,
  "goodRate": 99,
  "url": "https://item.jd.com/100309589315.html",
  "mobileUrl": "https://item.m.jd.com/product/100309589315.html",
  "image": "https://img14.360buyimg.com/n0/jfs/t1/549221/29/23/58615/6ad02b79Fe61dcfb3/0083320320c8ae56.jpg",
  "images": ["https://img14.360buyimg.com/n0/jfs/t1/549221/29/23/58615/6ad02b79Fe61dcfb3/0083320320c8ae56.jpg"],
  "categoryIds": ["9987", "653", "655"],
  "inStock": true,
  "keyword": null,
  "position": null,
  "totalResults": null,
  "input": "https://item.jd.com/100309589315.html",
  "mode": "product",
  "status": "complete",
  "error": null,
  "fetchedAt": "2026-10-15T09:56:28.416Z"
}

null means JD did not show this value; it is not 0. goodRate: null means "no reviews yet", not "0% positive". In a large test (20 keywords, 3,686 rows) 87% of rows had a review count and 86% a good rate; the rest were mostly new products.

Pricing and limits

Pricing unit

Pay per event. You pay for:

• Actor start — $0.005 per run (Apify counts one start per GB of memory, at least one; the default 256 MB is one);
• product — each saved complete or partial product row.

Event	Free	Bronze	Silver	Gold
Actor start	$0.005	$0.005	$0.005	$0.005
product	$0.003	$0.0029	$0.0028	$0.0027

Platform usage (compute, proxy) is included. Keywords with no products, not_found and error rows are free; you pay only the start fee for a run that saves no products. If a run fails, you pay only for the rows it saved. The run stops when it reaches your maximum charge and says so in RUN_SUMMARY (reason: CHARGE_LIMIT).

Cost examples (Free plan)

Scenario	Input scope	Rows	Charge	Stop/cap behavior
Prefill	1 keyword, 10 products	10	$0.035	about 4 s
Normal search	3 keywords × 50 products	150	$0.455
All products for a keyword	1 keyword × 1000	about 180	about $0.545	JD gives about 150–200
Product lookup	50 SKUs	50	$0.155	about 80 s; not_found and wrong links free
Nothing found	1 keyword with no products	0	$0.005	start fee only

Limits

• Up to 20 keywords per run, up to 1,000 products per keyword (JD gives about 150–200); up to 50 product links per run.
• Measured: 20 keywords × 1,000 → 3,686 rows in 200 seconds; 3 keywords → 461 rows in 27 seconds; 50 real product links → 74–89 seconds; 100 product links that do not exist → 31 seconds. For more products, split the list into runs of 50.
• The default run timeout is 300 seconds. The Actor stops on its own at about 240 seconds and keeps saved rows (reason: DEADLINE).

API, export and integrations

• API: start the Actor with the Apify API and read the Dataset items as JSON, CSV or Excel.
• Export: the Output tab exports JSON, CSV, Excel, XML and RSS.
• Scheduling: run the same keywords daily with Apify Schedules and compare price by skuId and fetchedAt.
• Webhooks / no-code: send results to Google Sheets, Make or Zapier through Apify integrations.

Use with AI agents (MCP)

• MCP server with only this Actor: https://mcp.apify.com?tools=cliqtomedia/jd-com-scraper
• Smallest input:

{ "keywords": ["手机"], "maxItems": 10 }

• Key output fields: title, price (CNY), shop, commentsCountMin, goodRate, url.
• 10 products take about 4 seconds and cost $0.035. Keep maxItems small for quick answers; JD gives at most about 200 products per keyword.

Troubleshooting

Fewer products than maxItems

Symptom: a keyword returns 150–200 rows when you asked for 1,000. Likely reason: JD's public search lists only about 150–200 products per keyword. The run message says how many keywords ran out. Check: RUN_SUMMARY.targets and the run message. Action: add more specific keywords (brand + product, for example 华为手机).

price is null in Product mode

Symptom: a row with status: partial and an error that starts with "Price not found". Likely reason: JD hides prices on product pages without a login, and this SKU did not appear in JD search for its title. Check: the error text. Action: use the title, shop and images, or search by a keyword that finds the product.

A keyword returns nothing

Symptom: no rows for a keyword; RUN_SUMMARY.status is partial or not_found. Likely reason: JD has no products for this word. Check: RUN_SUMMARY.targets — the keyword is not_found. Action: try the Chinese name of the product.

The run failed with SOURCE_ERROR

Symptom: RUN_SUMMARY.reason: SOURCE_ERROR; the run ends FAILED, even if most rows were saved. Likely reason: JD sent a verification page for some requests, or did not answer after several tries with new proxy IPs. Check: saved rows are in the Dataset and charged; the products or keywords JD did not answer have status: error in the Dataset and in RUN_SUMMARY.targets, and they are free. Action: run only those products or keywords again a bit later. This is a failure, not an empty result.

FAQ

Do I need a JD account, cookies or API key?

No. The Actor reads public JD pages without a login. It does not solve captchas or log in.

Can I get review texts, specifications or variants?

Not in this Actor. JD shows these only after a login. You get the review count and good rate.

Why are the prices different from the JD app?

The Actor sees the price JD shows to a visitor without an account. Member prices, coupons and app-only offers are not included.

Are the results complete and fresh?

The rows show what JD's public search shows at fetchedAt. JD changes order and prices often. The Actor does not promise every product for a keyword.

Related Actors

• Historical FX Rates Converter — convert CNY prices to your currency for a given date.
• Google Autocomplete Scraper — find product keywords to use in keywords.
• Website JSON-LD Structured Data Scraper — read product prices from other online stores.

Support

For bugs, questions or feature requests, open the Issues tab of this Actor. Add the run ID, your input and the error or RUN_SUMMARY message. Never send tokens, cookies or passwords.